Digital medicine and the curse of dimensionality
Open accessAbstract
Digital health data are multimodal and high-dimensional. A patient’s health state can be characterized by a multitude of signals including medical imaging, clinical variables, genome sequencing, conversations between clinicians and patients, and continuous signals from wearables, among others. This high volume, personalized data stream aggregated over patients’ lives has spurred interest in developing new artificial intelligence (AI) models for higher-precision diagnosis, prognosis, and tracking. While the promise of these algorithms is undeniable, their dissemination and adoption have been slow, owing partially to unpredictable AI model performance once deployed in the real world. We posit that one of the rate-limiting factors in developing algorithms that generalize to real-world scenarios is the very attribute that makes the data exciting—their high-dimensional nature. This paper considers how the large number of features in vast digital health data can challenge the development of robust AI models—a phenomenon known as “the curse of dimensionality” in statistical learning theory. We provide an overview of the curse of dimensionality in the context of digital health, demonstrate how it can negatively impact out-of-sample performance, and highlight important considerations for researchers and algorithm designers.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
There has not been enough publicly available audio or video examples of speech and language disorders to develop AI systems that work well for children or adults with these disorders.
“First, and perhaps most critically, there has not been enough publicly available audio or video examples of speech and language disorders to develop AI systems that work well for children or adults with these types of disorders (e.g., Benway et al., 2023;Berisha et al., 2021;Shahin et al., 2020).”
From Equipping Speech-Language Clinicians for the Critical Appraisal of an Artificial Intelligence–Driven, Evidence-Based Future (2025) · cites this paper as Berisha et al., 2021
The Introduction says clinical speech databases are often small, with tens or hundreds of patients and only a few minutes of speech per person, making speech-based clinical AI challenging. The cited paper does not establish that publicly available audio or video examples of speech and language disorders are insufficient, or specifically assess AI performance for children and adults with those disorders. (LLM, conf=0.86)
Experimental — accuracy not yet measuredThe more complex the association between the input and the output, the more data are needed to avoid overfitting and ensure a tool will work well in the future.
“The more complex the association between the input and the output, the more data are needed to avoid overfitting and ensure a tool will work well in the future (Berisha et al., 2021), which highlights the importance of machine learning experimental design for data-scarce clinical AI (Ghasemzadeh et al., 2024).”
From Equipping Speech-Language Clinicians for the Critical Appraisal of an Artificial Intelligence–Driven, Evidence-Based Future (2025) · cites this paper as Berisha et al., 2021
High dimensionality of data may be a rate-limiting factor in development of algorithms capable of generalizing to real-world scenarios.
“The high dimensionality of data that makes it attractive to research may, at the same time, be a rate-limiting factor in the development of algorithms capable of generalizing to real-world scenarios45.”
From Orchestrating explainable artificial intelligence for multimodal and longitudinal data in medical imaging (2024) · cites this paper as [45]
Dataset blind spots can hurt model performance in terms of real-life prediction and should be considered early in the model development and evaluation process.
“These dataset blind spots can hurt model performance in terms of real-life prediction and should therefore be considered early in the model development and evaluation process 123”
From Multimodal biomedical AI (2022) · cites this paper as [123]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Rethinking the Inception Architecture for Computer Vision
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.308 · Cited by 32,138 · see citing works
Why Most Published Research Findings Are False
Ioannidis, J. P. A. (2005) · PLoS Medicine · doi:10.1371/journal.pmed.0020124 · Cited by 10,505 · see citing works
High-performance medicine: the convergence of human and artificial intelligence
Topol, E. J. (2019) · Nature Medicine · doi:10.1038/s41591-018-0300-7 · Cited by 10,450 · see citing works
The file drawer problem and tolerance for null results.
Rosenthal, R. (1979) · Psychological Bulletin · doi:10.1037/0033-2909.86.3.638 · Cited by 9,068 · see citing works
Why Most Published Research Findings Are False
Ioannidis, J. P. A. (2005) · CHANCE · doi:10.1080/09332480.2005.10722754 · Cited by 4,702 · see citing works
Valiant, L. G. (1984) · ACM eBooks · doi:10.1145/3749740.3749746 · Cited by 4,211 · see citing works
Valiant, L. G. (1984) · Communications of the ACM · doi:10.1145/1968.1972 · Cited by 3,279 · see citing works
Heart rate variability: A review
Rajendra Acharya, U., Paul Joseph, K., Kannathal, N., Lim, C. M., & Suri, J. S. (2006) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-006-0119-0 · Cited by 2,820 · see citing works
Machine learning algorithm validation with a limited sample size
Vabalas, A., Gowen, E., Poliakoff, E., & Casson, A. J. (2019) · PLOS ONE · doi:10.1371/journal.pone.0224365 · Cited by 1,901 · see citing works
Statistics for High-Dimensional Data
Bühlmann, P., & van de Geer, S. (2011) · Springer series in statistics · doi:10.1007/978-3-642-20192-9 · Cited by 1,697 · see citing works
Statistics for High-Dimensional Data: Methods, Theory and Applications
Bhlmann, P., & van de Geer, S. (2011) · Cited by 1,527 · see citing works
Single subject prediction of brain disorders in neuroimaging: Promises and pitfalls
Arbabshirani, M. R., Plis, S., Sui, J., & Calhoun, V. D. (2017) · NeuroImage · doi:10.1016/j.neuroimage.2016.02.079 · Cited by 912 · see citing works
A Note on the Use of Principal Components in Regression
Jolliffe, I. T. (1982) · Journal of the Royal Statistical Society Series C (Applied Statistics) · doi:10.2307/2348005 · Cited by 856 · see citing works
Transfusion: Understanding Transfer Learning for Medical ImagingPreprint
Raghu, M., Zhang, C., Kleinberg, J. M., & Samy Bengio, A. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1902.07208 · Cited by 845 · see citing works
Regularizing Neural Networks by Penalizing Confident Output Distributions
Pereyra, G., Tucker, G., Chorowski, J., Kaiser, Ł., & Hinton, G. (2017) · arXiv · doi:10.48550/arxiv.1701.06548 · Cited by 655 · see citing works
A comparison of PCA, KPCA and ICA for dimensionality reduction in support vector machine
Cao, L., Chua, K. S., Chong, W. K., Lee, H. P., & Gu, Q. (2003) · Neurocomputing · doi:10.1016/s0925-2312(03)00433-8 · Cited by 588 · see citing works
Predicting sample size required for classification performance
Figueroa, R. L., Zeng‐Treitler, Q., Kandula, S., & Ngo, L. H. (2012) · BMC Medical Informatics and Decision Making · doi:10.1186/1472-6947-12-8 · Cited by 526 · see citing works
Maximum Performance Tests of Speech Production
Kent, R. D., Kent, J. F., & Rosenbek, J. C. (1987) · Journal of Speech and Hearing Disorders · doi:10.1044/jshd.5204.367 · Cited by 395 · see citing works
Language in the aging brain: The network dynamics of cognitive decline and preservation
Shafto, M. A., & TYLER, L. K. (2014) · Science · doi:10.1126/science.1254404 · Cited by 361 · see citing works
Boosting and Other Ensemble Methods
Drucker, H., Cortes, C., Jackel, L. D., LeCun, Y., & Vapnik, V. (1994) · Neural Computation · doi:10.1162/neco.1994.6.6.1289 · Cited by 357 · see citing works
The reusable holdout: Preserving validity in adaptive data analysis
Dwork, C., Feldman, V., Hardt, M., Pitassi, T., Reingold, O., & Roth, A. (2015) · Science · doi:10.1126/science.aaa9375 · Cited by 307 · see citing works
Multi-Task Self-Supervised Learning for Robust Speech Recognition
Ravanelli, M., Zhong, J., Pascual, S., Świętojański, P., Monteiro, J., Trmal, J., & Bengio, Y. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9053569 · Cited by 276 · see citing works
García, S. D. L. F., Ritchie, C. W., & Luz, S. F. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.06047 · Cited by 265 · see citing works
Preserving Statistical Validity in Adaptive Data Analysis
Dwork, C., Feldman, V., Hardt, M., Pitassi, T., Reingold, O., & Roth, A. (2015) · ACM Symposium on Theory of Computing (STOC) · doi:10.1145/2746539.2746580 · Cited by 258 · see citing works
Geographic Distribution of US Cohorts Used to Train Deep Learning Algorithms
Kaushal, A., Altman, R., & Langlotz, C. (2020) · JAMA · doi:10.1001/jama.2020.12067 · Cited by 239 · see citing works
Poeppel, D. (2012) · Cognitive Neuropsychology · doi:10.1080/02643294.2012.710600 · Cited by 233 · see citing works
Coveney, P. V., Dougherty, E. R., & Highfield, R. R. (2016) · Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences · doi:10.1098/rsta.2016.0153 · Cited by 229 · see citing works
Bulbar and speech motor assessment in ALS: Challenges and future directions
Green, J. R., Yunusova, Y., Kuruvilla-Dugdale, M., Wang, J., Pattee, G. L., Synhorst, L. L., Zinman, L. H., & Berry, J. D. (2013) · Amyotrophic Lateral Sclerosis and Frontotemporal Degeneration · doi:10.3109/21678421.2013.817585 · Cited by 221 · see citing works
Do CIFAR-10 Classifiers Generalize to CIFAR-10?Preprint
Recht, B., Roelofs, R., Schmidt, L., & Shankar, V. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.00451 · Cited by 218 · see citing works
Publication bias in meta‐analyses from the Cochrane Database of Systematic Reviews
Kiciński, M., Springate, D. A., & Kontopantelis, E. (2015) · Statistics in Medicine · doi:10.1002/sim.6525 · Cited by 213 · see citing works
On the Dangers of Cross-Validation. An Experimental Evaluation
Rao, R. B., Fung, G., & Rosales, R. (2008) · SIAM International Conference on Data Mining · doi:10.1137/1.9781611972788.54 · Cited by 207 · see citing works
Dialect variation and formant frequency: The American English vowels revisited
Hagiwara, R. (1997) · The Journal of the Acoustical Society of America · doi:10.1121/1.419712 · Cited by 187 · see citing works
A systematic literature review of automatic Alzheimer’s disease detection from speech and language
Petti, U., Baker, S., & Korhonen, A. (2020) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocaa174 · Cited by 168 · see citing works
Overlap in observational studies with high-dimensional covariates
D’Amour, A., Ding, P., Feller, A., Lei, L., & Sekhon, J. S. (2020) · Journal of Econometrics · doi:10.1016/j.jeconom.2019.10.014 · Cited by 155 · see citing works
Nearest Neighbour Searches and the Curse of Dimensionality
MARIMONT, R. B., & SHAPIRO, M. B. (1979) · IMA Journal of Applied Mathematics · doi:10.1093/imamat/24.1.59 · Cited by 146 · see citing works
Challenges in Personalized Nutrition and Health
Verma, M., Hontecillas, R., Tubau‐Juni, N., Abedi, V., & Bassaganya-Riera, J. (2018) · Frontiers in Nutrition · doi:10.3389/fnut.2018.00117 · Cited by 145 · see citing works
Mueller, K. D., Koscik, R. L., Hermann, B. P., Johnson, S. C., & Turkstra, L. S. (2018) · Frontiers in Aging Neuroscience · doi:10.3389/fnagi.2017.00437 · Cited by 128 · see citing works
Systematic misestimation of machine learning performance in neuroimaging studies of depression
Flint, C., Cearns, M., Opel, N., Redlich, R., Mehler, D. M. A., Emden, D., Winter, N. R., Leenings, R., Eickhoff, S. B., Kircher, T., et al. (2021) · Neuropsychopharmacology · doi:10.1038/s41386-021-01020-7 · Cited by 120 · see citing works
A practical Alzheimer’s disease classifier via brain imaging-based deep learning on 85,721 samples
Lu, B., Li, H., Chang, Z., Li, L., Chen, N.-X., Zhu, Z.-C., Zhou, H., Li, X., Wang, Y., Cui, S., et al. (2022) · Journal Of Big Data · doi:10.1186/s40537-022-00650-y · Cited by 105 · see citing works
Towards speaker adaptive training of deep neural network acoustic models
Miao, Y., Zhang, H., & Metze, F. (2014) · Interspeech (USB) · doi:10.21437/interspeech.2014-490 · Cited by 102 · see citing works
Effects of Aging on Vocal Fundamental Frequency and Vowel Formants in Men and Women
Eichhorn, J. T., Kent, R. D., Austin, D., et al. (2017) · Journal of Voice · doi:10.1016/j.jvoice.2017.08.003 · Cited by 95 · see citing works
Decision trees do not generalize to new variations
Bengio, Y., Delalleau, O., & Simard, C. (2010) · Computational Intelligence · doi:10.1111/j.1467-8640.2010.00366.x · Cited by 89 · see citing works
Empirically Estimable Classification Bounds Based on a Nonparametric Divergence Measure
Berisha, V., Wisler, A. A., Hero, A. O., & Spanias, A. (2015) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2015.2477805 · Cited by 76 · see citing works
Self-citationOn the Hardness of Domain Adaptation and the Utility of Unlabeled Target Samples
Ben-David, S., & Urner, R. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-34106-9_14 · Cited by 57 · see citing works
Repeatability of Commonly Used Speech and Language Features for Clinical Applications
Stegmann, G. M., Hahn, S., Liss, J., Shefner, J., Rutkove, S. B., Kawabata, K., Bhandari, S., Shelton, K., Duncan, C. J., & Berisha, V. (2020) · Digital Biomarkers · doi:10.1159/000511671 · Cited by 47 · see citing works
Self-citationFinger tapping impairments are highly sensitive for evaluating upper motor neuron lesions
Shirani, A., Newton, B. D., & Okuda, D. T. (2017) · BMC Neurology · doi:10.1186/s12883-017-0829-y · Cited by 37 · see citing works
Regularization via Structural Label SmoothingPreprint
Li, W., Dasarathy, G., & Berisha, V. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.01900 · Cited by 28 · see citing works
Self-citationMaley, J. H., Wanis, K. N., Young, J. G., & Celi, L. A. (2020) · BMJ Health & Care Informatics · doi:10.1136/bmjhci-2020-100220 · Cited by 19 · see citing works
de la Fuente Garcia, S., Ritchie, C. W., & Luz, S. (2020) · Journal of Alzheimer's Disease · doi:10.3233/jad-200888 · Cited by 5 · see citing works
Who does the model learn from?
Charpignon, M.-L., Celi, L. A., & Samuel, M. C. (2021) · The Lancet Digital Health · doi:10.1016/s2589-7500(21)00057-1 · Cited by 4 · see citing works
Flagged first, then by citations
Research works printed but not linked: 2
- [4]Koutroumbas, K. & Theodoridis, S. Pattern Recognition (4th Ed.). (Elsevier Inc., Burlington, 2009).
- [47]Goodfellow, I., Shlens, J. & Szegedy, C. Explaining and Harnessing Adversarial Examples. Proceedings of the International Conference on Learning Representations (2015).
Cited by
Sorted by citations
Acosta, J. N., Falcone, G. J., Rajpurkar, P., & Topol, E. J. (2022) · Nature Medicine · doi:10.1038/s41591-022-01981-2 · Cited by 1,365 · see citing works
Pahud de Mortanges, A., Luo, H., Shu, S. Z., Kamath, A., Suter, Y., Shelan, M., Pöllinger, A., & Reyes, M. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01190-w · Cited by 93 · see citing works
Ghasemzadeh, H., Hillman, R. E., & Mehta, D. D. (2024) · Journal of Speech, Language, and Hearing Research · doi:10.1044/2023_jslhr-23-00273 · Cited by 72 · see citing works
Berisha, V., & Liss, J. M. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01199-1 · Cited by 41 · see citing works
Self-citationPre-training in Medical Data: A Survey
Qiu, Y., Lin, F., Chen, W., & Xu, M. (2023) · Machine Intelligence Research · doi:10.1007/s11633-022-1382-8 · Cited by 30 · see citing works
Low, D. M., Rao, V., Randolph, G., Song, P. C., & Ghosh, S. S. (2024) · PLOS Digital Health · doi:10.1371/journal.pdig.0000516 · Cited by 16 · see citing works
Perfalk, E., Damgaard, J. G., Bernstorff, M., Hansen, L., Danielsen, A. A., & Østergaard, S. D. (2024) · Psychological Medicine · doi:10.1017/s0033291724002642 · Cited by 12 · see citing works
Benway, N. R., & Preston, J. L. (2025) · Language, Speech, and Hearing Services in Schools · doi:10.1044/2025_lshss-24-00085 · Cited by 12 · see citing works
- Authors
- Berisha, V., Krantsevich, C., Hahn, P. R., Hahn, S., Dasarathy, G., Turaga, P., & Liss, J.
- Affiliations
- Arizona State University (US)
- Corresponding author
- Berisha, Visar
- Venue
- npj Digital Medicine
- Published
- 2021-10-28
- Volume
- 4
- Issue
- 1
- Pages
- 153-153
- ISSN
- 2398-6352 (ISSN-L)
- PMC
- PMC8553745
- OpenAlex
- W3208504364
- Cited by
- 344
- References
- 50
- Citations by year
- 2026: 68 · 2025: 92 · 2024: 89 · 2023: 65 · 2022: 30
- FWCI
- 8.16
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- npj Digital Medicine (publishedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, other-oa)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- U.S. Department of Health & Human Services | National Institutes of Health (5R01DC006859)
- United States Department of Defense | United States Navy | Office of Naval Research (N00014-17-1-2826)
- U.S. Department of Health & Human Services | NIH | National Institute on Deafness and Other Communication Disorders
- Topics
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- curse of dimensionality, digital health data, high-dimensional data, artificial intelligence models, out-of-sample generalization, model robustness, multimodal health data, clinical decision support, wearable device data, diagnostic imaging, clinical factors, genome sequencing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
